Jonas Schievink [Thu, 26 Nov 2020 12:39:09 +0000 (13:39 +0100)]
Rollup merge of #79365 - richkadel:llvm-cov-map-version-4, r=wesleywiser
Upgrades the coverage map to Version 4
Changes the coverage map injected into binaries compiled with
`-Zinstrument-coverage` to LLVM Coverage Mapping Format, Version 4 (from
Version 3). Note, binaries compiled with this version will require LLVM
tools from at least LLVM Version 11.
Jonas Schievink [Thu, 26 Nov 2020 12:39:05 +0000 (13:39 +0100)]
Rollup merge of #79000 - sivadeilra:user/ardavis/lev_distance, r=wesleywiser
Move lev_distance to rustc_ast, make non-generic
rustc_ast currently has a few dependencies on rustc_lexer. Ideally, an AST
would not have any dependency its lexer, for minimizing
design-time dependencies. Breaking this dependency would also have practical
benefits, since modifying rustc_lexer would not trigger a rebuild of rustc_ast.
This commit does not remove the rustc_ast --> rustc_lexer dependency,
but it does remove one of the sources of this dependency, which is the
code that handles fuzzy matching between symbol names for making suggestions
in diagnostics. Since that code depends only on Symbol, it is easy to move
it to rustc_span. It might even be best to move it to a separate crate,
since other tools such as Cargo use the same algorithm, and have simply
contain a duplicate of the code.
This changes the signature of find_best_match_for_name so that it is no
longer generic over its input. I checked the optimized binaries, and this
function was duplicated for nearly every call site, because most call sites
used short-lived iterator chains, generic over Map and such. But there's
no good reason for a function like this to be generic, since all it does
is immediately convert the generic input (the Iterator impl) to a concrete
Vec<Symbol>. This has all of the costs of generics (duplicated method bodies)
with no benefit.
Changing find_best_match_for_name to be non-generic removed about 10KB of
code from the optimized binary. I know it's a drop in the bucket, but we have
to start reducing binary size, and beginning to tame over-use of generics
is part of that.
Jonas Schievink [Thu, 26 Nov 2020 12:39:00 +0000 (13:39 +0100)]
Rollup merge of #77758 - Stupremee:turbofish-help-for-const, r=varkor
suggest turbofish syntax for uninferred const arguments
When not providing a const generic value, and it can not be inferred, the following suggestion is suggested:
![image](https://user-images.githubusercontent.com/39732259/95616180-af127b80-0a69-11eb-8877-551c815f9627.png)
bors [Wed, 25 Nov 2020 21:22:46 +0000 (21:22 +0000)]
Auto merge of #79411 - tmiasko:naked-params, r=Amanieu
Validate use of parameters in naked functions
* Reject use of parameters inside naked function body.
* Reject use of patterns inside function parameters, to emphasize role
of parameters a signature declaration (mirroring existing behaviour
for function declarations) and avoid generating code introducing
specified bindings.
Closes issues below by considering input to be ill-formed.
bors [Wed, 25 Nov 2020 14:08:47 +0000 (14:08 +0000)]
Auto merge of #79400 - camelid:rustdoc-summary_opts, r=GuillaumeGomez
Add `summary_opts()` for Markdown summary rendering options
We have a similar function `opts()` that is for rendering the main body
of the documentation, but until now we just constructed the options for
rendering summaries on the fly. This is a problem if/when we change the
enabled options since the different places can get out-of-sync.
bors [Wed, 25 Nov 2020 07:25:19 +0000 (07:25 +0000)]
Auto merge of #79336 - camelid:rename-feature-oibit-to-auto, r=oli-obk
Rename `optin_builtin_traits` to `auto_traits`
They were originally called "opt-in, built-in traits" (OIBITs), but
people realized that the name was too confusing and a mouthful, and so
they were renamed to just "auto traits". The feature flag's name wasn't
updated, though, so that's what this PR does.
There are some other spots in the compiler that still refer to OIBITs,
but I don't think changing those now is worth it since they are internal
and not particularly relevant to this PR.
Also see <https://rust-lang.zulipchat.com/#narrow/stream/131828-t-compiler/topic/opt-in.2C.20built-in.20traits.20(auto.20traits).20feature.20name>.
r? `@oli-obk` (feel free to re-assign if you're not the right reviewer for this)
bors [Wed, 25 Nov 2020 05:10:53 +0000 (05:10 +0000)]
Auto merge of #77491 - lukaslueg:peek_mut, r=m-ou-se
Proposal to add Peekable::peek_mut
A "peekable" iterator has a `peek()`-method which provides an immutable reference to the next item. We currently do not have a method to modify that item, which we could easily add via a `peek_mut()`. See the test for a use-case (alike to my original use case), where a "pristine" iterator is passed on after modifying its state via `peek_mut()`.
If there is interest in this, I can expand on the tests and docs.
Camelid [Wed, 25 Nov 2020 03:54:41 +0000 (19:54 -0800)]
Add `summary_opts()` for Markdown summary rendering options
We have a similar function `opts()` that is for rendering the main body
of the documentation, but until now we just constructed the options for
rendering summaries on the fly. This is a problem if/when we change the
enabled options since the different places can get out-of-sync.
bors [Wed, 25 Nov 2020 02:49:28 +0000 (02:49 +0000)]
Auto merge of #76688 - yokodake:patch-2, r=kodrAus
Document unsafety in core::slice::memchr
Contributes to #66219
Note sure if that's good enough, especially for the `align_to` call.
The docs only mention transmuting and I don't think that everything related to reference lifetimes and state validity mentioned in the [nomicon](https://doc.rust-lang.org/nomicon/transmutes.html) are relevant here.
Arlie Davis [Thu, 12 Nov 2020 19:24:10 +0000 (11:24 -0800)]
Move lev_distance to rustc_ast, make non-generic
rustc_ast currently has a few dependencies on rustc_lexer. Ideally, an AST
would not have any dependency its lexer, for minimizing unnecessarily
design-time dependencies. Breaking this dependency would also have practical
benefits, since modifying rustc_lexer would not trigger a rebuild of rustc_ast.
This commit does not remove the rustc_ast --> rustc_lexer dependency,
but it does remove one of the sources of this dependency, which is the
code that handles fuzzy matching between symbol names for making suggestions
in diagnostics. Since that code depends only on Symbol, it is easy to move
it to rustc_span. It might even be best to move it to a separate crate,
since other tools such as Cargo use the same algorithm, and have simply
contain a duplicate of the code.
This changes the signature of find_best_match_for_name so that it is no
longer generic over its input. I checked the optimized binaries, and this
function was duplicated at nearly every call site, because most call sites
used short-lived iterator chains, generic over Map and such. But there's
no good reason for a function like this to be generic, since all it does
is immediately convert the generic input (the Iterator impl) to a concrete
Vec<Symbol>. This has all of the costs of generics (duplicated method bodies)
with no benefit.
Changing find_best_match_for_name to be non-generic removed about 10KB of
code from the optimized binary. I know it's a drop in the bucket, but we have
to start reducing binary size, and beginning to tame over-use of generics
is part of that.
Tomasz Miąsko [Wed, 25 Nov 2020 00:00:00 +0000 (00:00 +0000)]
Validate use of parameters in naked functions
* Reject use of parameters inside naked function body.
* Reject use of patterns inside function parameters, to emphasize role
of parameters a signature declaration (mirroring existing behaviour
for function declarations) and avoid generating code introducing
specified bindings.
Rich Kadel [Tue, 24 Nov 2020 19:50:24 +0000 (11:50 -0800)]
Check for LLVM 11+ when using `-Z instrument-coverage`
* `rustc` should now compile under LLVM 9 or 10
* Compiler generates an error if `-Z instrument-coverage` is specified
but LLVM version is less than 11
* Coverage tests that require `-Z instrument-coverage` and run codegen
should be skipped if LLVM version is less than 11
bors [Tue, 24 Nov 2020 18:32:19 +0000 (18:32 +0000)]
Auto merge of #79384 - ehuss:update-cargo, r=ehuss
Update cargo
10 commits in 2af662e22177a839763ac8fb70d245a680b15214..bfca1cd22bf514d5f2b6c1089b0ded0ba7dfaa6e
2020-11-12 19:04:56 +0000 to 2020-11-24 16:33:21 +0000
- Shrink the progress bar, to give more space after it. (rust-lang/cargo#8892)
- Add some comments to the toml code (rust-lang/cargo#8887)
- Start searching git config at new path (rust-lang/cargo#8886)
- Fix documentation for CARGO_PRIMARY_PACKAGE. (rust-lang/cargo#8891)
- Bump to 0.51.0, update changelog (rust-lang/cargo#8894)
- Publish target's "doc" setting when emitting metadata (rust-lang/cargo#8869)
- Relaxes expectation of `cargo test` tests to accept test execution time (rust-lang/cargo#8884)
- Finish implementation of `-Zextra-link-arg`. (rust-lang/cargo#8441)
- Reproducible crate builds (rust-lang/cargo#8864)
- Allow resolver="1" to explicitly use the old resolver behavior. (rust-lang/cargo#8857)
bors [Tue, 24 Nov 2020 12:27:30 +0000 (12:27 +0000)]
Auto merge of #79377 - jonas-schievink:rollup-ye81i66, r=jonas-schievink
Rollup of 10 pull requests
Successful merges:
- #76858 (Add exploit mitigations chapter to the rustc book)
- #79310 (Make `fold_item_recur` non-nullable)
- #79312 (Get rid of `doctree::Impl`)
- #79321 (Accept '!' in intra-doc links)
- #79346 (Allow using `-Z fewer-names=no` to retain value names)
- #79351 (Fix typo in `keyword` docs for traits)
- #79354 (BTreeMap: cut out the ceremony around BoxedNode)
- #79358 (BTreeMap/BTreeSet: make public doc more consistent)
- #79367 (Allow disabling TrapUnreachable via -Ztrap-unreachable=no)
- #79374 (Add note to use nightly when using expr in const generics)
Jonas Schievink [Tue, 24 Nov 2020 12:17:51 +0000 (13:17 +0100)]
Rollup merge of #79374 - mendess:const-param-expr-diagnostic, r=lcnr
Add note to use nightly when using expr in const generics
As recommended by `@Icnr` in #73899 and in zulip, I've added a note saying that const expressions can be used in nightly.
```
error: generic parameters may not be used in const operations
--> $DIR/issue-61935.rs:10:23
|
6 | Self:FooImpl<{N==0}>
| ^ cannot perform const operation using `N`
|
= help: const parameters may only be used as standalone arguments, i.e. `N`
= note: use feature(const_generics) and feature(const_evaluatable_checked) to enable this
Jonas Schievink [Tue, 24 Nov 2020 12:17:49 +0000 (13:17 +0100)]
Rollup merge of #79367 - Dirbaio:trap-unreachable, r=jonas-schievink
Allow disabling TrapUnreachable via -Ztrap-unreachable=no
Currently this is only possible by defining a custom target, which is quite unwieldy.
This is useful for embedded targets where small code size is desired. For example, on my project (thumbv7em-none-eabi) this yields a 0.6% code size reduction: 132892 bytes -> 132122 bytes (770 bytes down).
Jonas Schievink [Tue, 24 Nov 2020 12:17:41 +0000 (13:17 +0100)]
Rollup merge of #79346 - tmiasko:more-names, r=jonas-schievink
Allow using `-Z fewer-names=no` to retain value names
Change `-Z fewer-names` into an optional boolean flag and allow using it
to either discard value names when true or retain them when false,
regardless of other settings.
Jonas Schievink [Tue, 24 Nov 2020 12:17:37 +0000 (13:17 +0100)]
Rollup merge of #79312 - jyn514:doctree-impl, r=GuillaumeGomez
Get rid of `doctree::Impl`
Follow-up to https://github.com/rust-lang/rust/pull/79264, continues breaking up https://github.com/rust-lang/rust/pull/78082. At some point I want to introduce `MaybeInlined`, but I think I'll wait until I need `MaybeInlined::InlinedWithOriginal` because it's not very useful in other situations.
Jonas Schievink [Tue, 24 Nov 2020 12:17:35 +0000 (13:17 +0100)]
Rollup merge of #79310 - jyn514:fold-item-cleanup, r=GuillaumeGomez
Make `fold_item_recur` non-nullable
This gets rid of a bunch of `unwrap()`s and makes it a little more clear
what's going on.
Originally I wanted to make `fold_item` non-nullable too, which would
have been a lot nicer to work with, but unfortunately `stripper` does
actually return `None` in some places. I might make a follow-up moving
stripper to be special and not a pass so that passes can be
non-nullable.
Found while working on https://github.com/rust-lang/rust/issues/76998.
Jonas Schievink [Tue, 24 Nov 2020 12:17:28 +0000 (13:17 +0100)]
Rollup merge of #76858 - rcvalle:rust-lang-exploit-mitigations, r=steveklabnik
Add exploit mitigations chapter to the rustc book
This section documents the exploit mitigations applicable to the Rust compiler when building programs for the Linux operating system on the AMD64 architecture and equivalent. This was done for a project I'm currently working on, and I hope it'll also be helpful to others.
bors [Tue, 24 Nov 2020 09:17:33 +0000 (09:17 +0000)]
Auto merge of #79294 - petrochenkov:determ, r=varkor
resolve: Do not put macros into `module.unexpanded_invocations` unless necessary
Macro invocations in modules <sup>(*)</sup> need to be tracked because they can produce named items when expanded.
We cannot give definite answer to queries like "does this module declare name `n`?" until all macro calls in that module are expanded.
Previously we marked too many macros as potentially producing named items.
E.g. in this example
```rust
mod m {
const C: u32 = line!();
}
```
`line!()` cannot emit any items into module `m`, but it was still marked.
This PR fixes that and marks macro calls as "unexpanded in module" only if they can actually emit named items into that module.
Diagnostics in UI test outputs have different order now because this change affects macro expansion order.
<sup>*</sup> Any containers for named items are called modules in resolve (that includes blocks, traits and enums in addition to `mod` items).
Ramon de C Valle [Fri, 18 Sep 2020 00:47:56 +0000 (17:47 -0700)]
Add exploit mitigations chapter to the rustc book
This section documents the exploit mitigations applicable to the Rust
compiler when building programs for the Linux operating system on the
AMD64 architecture and equivalent.
bors [Tue, 24 Nov 2020 00:30:25 +0000 (00:30 +0000)]
Auto merge of #78636 - dtolnay:puncteq, r=petrochenkov
Add PartialEq<char> for proc_macro::Punct
`punct.as_char() == '░'` is pervasive when parsing anything involving punct. I think `punct == '░'` is sufficiently unambiguous that it makes sense to provide the impl.
Allow disabling TrapUnreachable via -Ztrap-unreachable=no
This is useful for embedded targets where small code size is desired.
For example, on my project (thumbv7em-none-eabi) this yields a 0.6% code size reduction.
Rich Kadel [Mon, 23 Nov 2020 20:56:07 +0000 (12:56 -0800)]
Upgrades the coverage map to Version 4
Changes the coverage map injected into binaries compiled with
`-Zinstrument-coverage` to LLVM Coverage Mapping Format, Version 4 (from
Version 3). Note, binaries compiled with this version will require LLVM
tools from at least LLVM Version 11.
Camelid [Mon, 23 Nov 2020 03:54:31 +0000 (19:54 -0800)]
Rename `optin_builtin_traits` to `auto_traits`
They were originally called "opt-in, built-in traits" (OIBITs), but
people realized that the name was too confusing and a mouthful, and so
they were renamed to just "auto traits". The feature flag's name wasn't
updated, though, so that's what this PR does.
There are some other spots in the compiler that still refer to OIBITs,
but I don't think changing those now is worth it since they are internal
and not particularly relevant to this PR.
Also see <https://rust-lang.zulipchat.com/#narrow/stream/131828-t-compiler/topic/opt-in.2C.20built-in.20traits.20(auto.20traits).20feature.20name>.
bors [Mon, 23 Nov 2020 19:01:19 +0000 (19:01 +0000)]
Auto merge of #78439 - lzutao:rm-clouldabi, r=Mark-Simulacrum
Drop support for all cloudabi targets
`cloudabi` is a tier-3 target, and [it is no longer being maintained upstream][no].
This PR drops supports for cloudabi targets. Those targets are:
* aarch64-unknown-cloudabi
* armv7-unknown-cloudabi
* i686-unknown-cloudabi
* x86_64-unknown-cloudabi
Since this drops supports for a target, I'd like somebody to tag `relnotes` label to this PR.
Some other issues:
* The tidy exception for `cloudabi` crate is still remained because
* `parking_lot v0.9.0` and `parking_lot v0.10.2` depends on `cloudabi v0.0.3`.
* `parking_lot v0.11.0` depends on `cloudabi v0.1.0`.
bors [Mon, 23 Nov 2020 16:33:03 +0000 (16:33 +0000)]
Auto merge of #79345 - jonas-schievink:rollup-1yhhzx9, r=jonas-schievink
Rollup of 10 pull requests
Successful merges:
- #76829 (stabilize const_int_pow)
- #79080 (MIR visitor: Don't treat debuginfo field access as a use of the struct)
- #79236 (const_generics: assert resolve hack causes an error)
- #79287 (Allow using generic trait methods in `const fn`)
- #79324 (Use Option::and_then instead of open-coding it)
- #79325 (Reduce boilerplate with the `?` operator)
- #79330 (Fix typo in comment)
- #79333 (doc typo)
- #79337 (Use Option::map instead of open coding it)
- #79343 (Add my (`@flip1995)` work mail to the mailmap)
This works by threading const-ness of trait predicates through trait selection, in particular through `ParamCandidate`, and exposing it in the resulting `ImplSource`.
Since this change makes two bounds `T: Trait` and `T: ?const Trait` that only differ in their const-ness be treated like different bounds, candidate winnowing has been changed to drop the `?const` candidate in favor of the const candidate, to avoid ambiguities when both a const and a non-const bound is present.
Jonas Schievink [Mon, 23 Nov 2020 14:25:42 +0000 (15:25 +0100)]
Rollup merge of #79236 - lcnr:mcg-resolve-dsb, r=eddyb
const_generics: assert resolve hack causes an error
prevent the min_const_generics `HACK`s in resolve from triggering a fallback path which successfully compiles so that we don't have to worry about future compat issues when removing it
bors [Mon, 23 Nov 2020 14:20:22 +0000 (14:20 +0000)]
Auto merge of #79186 - JulianKnodt:str_from, r=Mark-Simulacrum
Change slice::to_vec to not use extend_from_slice
I saw this [Zulip thread](https://rust-lang.zulipchat.com/#narrow/stream/219381-t-libs/topic/String.3A.3Afrom%28.26str%29.20wonky.20codegen/near/216164455), and didn't see any update from it, so I thought I'd try to fix it. This converts `to_vec` to no longer use `extend_from_slice`, but relies on knowing that the allocated capacity is the same size as the input.
[Godbolt new v1](https://rust.godbolt.org/z/1bcWKG)
[Godbolt new v2 w/ drop guard](https://rust.godbolt.org/z/5jn76K)
[Godbolt old version](https://rust.godbolt.org/z/e4ePav)
After some amount of iteration, there are now two specializations for `to_vec`, one for `Copy` types that use memcpy, and one for clone types which is the original from this PR.
This is then used inside of `impl<T: Clone> FromIterator<Iter::Slice<T>> for Vec<T>` which is essentially equivalent to `&[T] -> Vec<T>`, instead of previous specialization of the `extend` function. This is because extend has to reason more about existing capacity by calling `reserve` on an existing vec, and thus produces worse asm.
Downsides: This allocates the exact capacity, so I think if many items are added to this `Vec` after, it might need to allocate whereas extending may not. I also noticed the number of faults went up in the benchmarks, but not sure where from exactly.
bors [Mon, 23 Nov 2020 07:00:30 +0000 (07:00 +0000)]
Auto merge of #77893 - petertodd:2020-impl-default-for-phantompinned, r=dtolnay
Impl Default for PhantomPinned
`PhantomPinned` is just a marker type, with an obvious default value (the only value). So I can't think of a reason not to do this. Sure, it's used in exotic situations with unsafe code. But the people writing that code can decide for themselves if they can derive `Default`, and in many situations the derived impl will make sense:
bors [Mon, 23 Nov 2020 04:47:25 +0000 (04:47 +0000)]
Auto merge of #76226 - CDirkx:const-ipaddr, r=dtolnay
Stabilize `IpAddr::is_ipv4` and `is_ipv6` as const
Insta-stabilize the methods `is_ipv4` and `is_ipv6` of `std::net::IpAddr` as const, in the same way as [PR#76198](https://github.com/rust-lang/rust/pull/76198).
Possible because of the recent stabilization of const control flow.
Tomasz Miąsko [Mon, 23 Nov 2020 00:00:00 +0000 (00:00 +0000)]
Allow using `-Z fewer-names=no` to retain value names
Change `-Z fewer-names` into an optional boolean flag and allow using it
to either discard value names when true or retain them when false,
regardless of other settings.